The Importance of Population-Based Estimates of Melanocytic Pathology

In this issue of JAMA Dermatology, Lott et al describe a novel and innovative approach to estimating the prevalence of different types of melanocytic lesions in all adults undergoing skin biopsies in a population. Using an automated natural language processing tool, they analyzed 80 368 written pathology reports of skin biopsies from 47 529 adult patients drawn from an underlying health system patient population, which was representative of the general adult population. For analysis of each pathology report, the diagnosis was first dichotomized as either a melanocytic or nonmelanocytic lesion. For melanocytic lesions, the cases were then assigned to 1 of 4 diagnostic categories based on the Melanocytic Pathology Assessment Tool and Hierarchy for Diagnosis (MPATH-Dx), with differing management implications: class I, benign lesion requiring no further treatment; class II, low-risk lesion requiring complete excision with narrow (<5-mm) margins; class III, higher-risk lesion such as melanoma in situ requiring reexcision with 5-mm to 1-cm margins; and class IV/V, invasive melanoma requiring wide reexcision with 1-cm margins or greater. Approximately one-quarter of all skin biopsies were of melanocytic lesions. Of these, over 90% were benign or low risk (class I or II); 4.5% were melanoma in situ or similar-risk lesion (class III); and 4.1% were invasive melanoma (class IV/V).